Genomic prediction using imputed whole-genome sequence data in Holstein Friesian cattle

Abstract

BackgroundIn contrast to currently used single nucleotide polymorphism (SNP) panels, the use of whole-genome sequence data is expected to enable the direct estimation of the effects of causal mutations on a given trait. This could lead to higher reliabilities of genomic predictions compared to those based on SNP genotypes. Also, at each generation of selection, recombination events between a SNP and a mutation can cause decay in reliability of genomic predictions based on markers rather than on the causal variants. Our objective was to investigate the use of imputed whole-genome sequence genotypes versus high-density SNP genotypes on (the persistency of) the reliability of genomic predictions using real cattle data.MethodsHighly accurate phenotypes based on daughter performance and Illumina BovineHD Beadchip genotypes were available for 5503 Holstein Friesian bulls. The BovineHD genotypes (631,428 SNPs) of each bull were used to impute whole-genome sequence genotypes (12,590,056 SNPs) using the Beagle software. Imputation was done using a multi-breed reference panel of 429 sequenced individuals. Genomic estimated breeding values for three traits were predicted using a Bayesian stochastic search variable selection (BSSVS) model and a genome-enabled best linear unbiased prediction model (GBLUP). Reliabilities of predictions were based on 2087 validation bulls, while the other 3416 bulls were used for training.ResultsPrediction reliabilities ranged from 0.37 to 0.52. BSSVS performed better than GBLUP in all cases. Reliabilities of genomic predictions were slightly lower with imputed sequence data than with BovineHD chip data. Also, the reliabilities tended to be lower for both sequence data and BovineHD chip data when relationships between training animals were low. No increase in persistency of prediction reliability using imputed sequence data was observed.ConclusionsCompared to BovineHD genotype data, using imputed sequence data for genomic prediction produced no advantage. To investigate the putative advantage of genomic prediction using (imputed) sequence data, a training set with a larger number of individuals that are distantly related to each other and genomic prediction models that incorporate biological information on the SNPs or that apply stricter SNP pre-selection should be considered.Electronic supplementary materialThe online version of this article (doi:10.1186/s12711-015-0149-x) contains supplementary material, which is available to authorized users.

Sequence variants selected from a multi-breed GWAS can improve the reliability of genomic predictions in dairy cattle

“…Several studies have shown that, using full sequence data directly for genomic prediction, rather than preselecting variants, does not improve prediction reliability [14, 15]. Our results show that both prediction reliability and genomic correlations across populations and breeds are highly sensitive to the choice of the prediction markers.…”

“…Bayesian variable selection models allow for heterogeneous variances and could potentially exploit the presence of causative mutations in the sequence data by assigning non-zero effects to variants that are close to the causative mutations, and zero effects to all other variants. However, in practice Van Binsbergen et al [15] found no increase in prediction reliability using full sequence data compared to SNPs, even with a Bayesian variable selection model. A potential explanation for the lack of improvement in prediction reliability could be that the number of SNPs is much larger than the number of individuals.…”

“…While whole-genome sequence data contain causative mutations and variants in high LD with some causative mutations, most of the variants are in low LD with the causative mutations. Thus, it is not surprising, that the use of whole-genome sequence data for genomic prediction does not necessarily increase reliabilities of genomic predictions compared to the use of genome-wide SNPs [14, 15], especially if the models used do not allow for sufficiently different within-breed variances and across-breed covariances for different SNPs.…”

Genotype Imputation Methods and Their Effects on Genomic Predictions in Cattle

“…The imputed HD and actual 50K SNP data yielded similar accuracies under all three methods [73]. Binsbergen et al [74] also reported no improvement in accuracy of genomic prediction was observed when using imputed sequence data over BovineHD data, suggesting that increases in density of genotypes may not necessarily lead to an increase in accuracy of genomic prediction with the current SNP panel information and statistical methods [75]. Previous studies [49,76,77] have shown evidence that RFI is a complex trait likely to be controlled by many SNPs with small effects.…”

Genomic prediction using preselected DNA variants from a GWAS with whole-genome sequence data in Holstein–Friesian cattle

“…Because most of the causal mutations that underlie quantitative trait loci (QTL) are expected to be included as genetic markers in the sequence data, it is expected that causal mutations will be identified more precisely than with the common lower density SNP chips [4] and that the reliability of genomic predictions and its persistency across generations and even across breeds [5, 6] will improve. This was confirmed on simulated data [7], but in practice, the use of cattle and chicken sequence data has not increased the reliability of genomic predictions [8, 9]. …”